sábado, marzo 21, 2009

Just some links

There are some nice things on line lately :).

First, Rod Page talk at NHM of London, he puts his slide-show on line, there is also a video and a pod-cast of the talk (I do not see them :P). The slide-show can give a nice idea about the talk. There are two important things on the talk: (a) the importance of the availability of data! I just agree with him: data must be freely available, and previously published data must be easy accessible; and (b) Scientific data must be readily usable, for example the “Encyclopedia of life” is just a fun site just like wikipedia or the web tree of life, they are full of nice pics and info, but cientifically irrelevant: no hard data are attached to it (i.e. morfological info, character matrices), this contrast with the most simpler ispaces of Page!

Although I do not like some of the comments, Malte Ebach made a good point against the revival of a “pragmatic” classification (for horticulture, for example). I think that the only rule of classification is the phylogeny. If an immutable classification is the objective, the only way I can thing is an “alphabetic” or “numeric” system, but the preservation of “traditional” names is not the solution. Maybe, it is better a change of the rules...

And speaking of change of the rules, I recommend the continuos checking of several of Mike Keesey posts. I do not think that phylocode is the solution (in fact I think that it is even worse than traditional Linnean classification), but he writes interesting things about databasing!

Also, it is seems to change some rules about the publication of names for taxonomy (via Evolving thoughts). I thing that only peer reviewed publication must count, at least from the last fifty years or so. Also I thing that new names, at least for species must be published on journals instead of books...

2 comentarios:

Mike Keesey dijo...

¡Gracias para la noticia!

1. "I think that the only rule of classification is the phylogeny."
2. "I do not think that phylocode is the solution (in fact I think that it is even worse than traditional Linnean classification)"

Without that parenthetical comment, you'd be in the company of some, like Paul Sereno and Jason Anderson, who support phylogenetic nomenclature, but not the PhyloCode. But instead, you seem to be simultaneously advocating phylogenetic nomenclature (#1), and rejecting it (#2). I'm not clear on how those two statements are mutually tenable. Care to elaborate?

If you'd like, you can join an appropriate discussion here: PhyloNom: What is it that you hate about the PhyloCode?

Note that my project, Names on Nodes, employs both rank-based nomenclature and phylogenetic nomenclature, and does not restrict the latter to the PhyloCode. Although I personally think the PhyloCode is an excellent idea and a well-designed code, Names on Nodes will utilize a variety of nomenclatural authorities.

Salva dijo...

I think that any statement of a classification must be supported by a phylogenetic analysis. For example, when someone proposes a new genus, this would be, ideally, accompanied by a phylogenetic analysis that support that taxonomic position (In true life, many taxonomist just propose the new genus, without a phylogentic analysis, which might render another genus as a paraphyletic unit, this is typical in cases where an species is found to be "aberrant", that is with many autapomorphies that "allow it" to be recognized as a genus).

But I'm a "strict cladist" xD, so any statement about a group is necessarily a statement about the characters that support the group. Then instead of a topology, the evidence for the placement continued to be the characters. That why the topology alone can not be a basis of classification (which is the core of most phyloCode taxonomy).

It can be said that I prefer "apomorphy definition". But although only synapomorphies provide evidence of grouping, also non apomorphic characters are implied by a group (for example, although it can not be used as evidence of grouping, the eggs of monotremates is a good character of this group). That is classification is about predicting characters in the members of a group!

But also, a name is about the coupling of the content and the evidence. Then, if the "apomorphic character" moves down on the tree, but the group content does not change, there is no reason to change the content of the group.
This is an example: Trichophora are a group of Pentatomomorpha bugs (Heteroptera) that have abdominal trichobothria. There are some Cimicomorpha (the sister group of Pentatomorpha) that also have abdominal trichobothria. Some authors (e.g. Carver et al., 1991; Schuh et al., 2009) argue that both trichobothria are homologs, then, it is possible than abdominal trichobothria is a synapomorphy of Cimico- and Pentatomomorpha, but the original Trichophora does not change, so although the name is actually "misleading", it continue to refer to the same group (of course supported by other characters). Curiously the heteropterous wing is not a synapomorphy of Heteroptera.

There are some problems of Linnean taxonomy (for me, the most important one is their inability to reject names, this characteristic that it is also shared under, at least as originally proposed, phyloCode taxonomy). But I think that coupled with some simple rules (such as Wiley proposals) can provide a useful taxonomy coupled with a phylogeny!